Gayana Bot. 72(2):72(2), 2015250-257, 2015 ISSN 0016-5301
Frugivory of Persea lingue (Lauraceae) and its effect on seed germination in southern Chile
Consumidores de frutos de Persea lingue (Lauraceae) en el sur de Chile y su efecto en la germinación de semillas
DAFNE GHO-ILLANES1*, CECILIA SMITH-RAMÍREZ1, 2, INAO A. VÁSQUEZ2 & IVÁN DÍAZ2
1Instituto de Ecología y Biodiversidad (IEB), Las Palmeras 3425, Ñuñoa, Santiago de Chile. 2Instituto de Conservación, Biodiversidad y Territorio, Facultad de Ciencias Forestales y Recursos Naturales, Campus Isla Teja, Universidad Austral de Chile, Valdivia, Chile. *[email protected]
ABSTRACT
Frugivory is a common interaction mechanism between plants and animals in the temperate rainforest of southern South America. In this study, we examined primary dispersal and predation stages of Persea lingue (Ruiz et Pav.) Nees (Lauraceae), an endemic large-seeded tree of these forests. The main goal was to determine the identity of vertebrate seed dispersers of P. lingue, to distinguish legitimate seed dispersers, pulp consumers and seed predators, and their influence on seed germination. In order to know which birds and small mammals were feeding P. lingue seeds, we observed 20 P. lingue canopies, and live traps baited with P. lingue seeds were installed in fragment forests in south-central Chile. To determine whether birds could be legitimate dispersers of P. lingue, we collected regurgitated and fecal samples. We examined whether rodents removed the fruit pulp, the seeds, or the whole fruit. Additionally, we studied if seeds placed on the forest floor were removed by rodents. In a final experiment we conducted germination trials to establish whether pulp removal (by birds, rodents, or artificially-removed) influenced germination rates. We found six bird species feeding on P. lingue seeds. None of them defecated intact seeds, but, Turdus falcklandii Quoy & Gaimard (Turdinae) and Xolmis pyrope (Kittlitz) (Tyrannidae) regurgitated intact seeds. We captured five species of rodents in Sherman traps. Four of them left intact seeds. We did not find any effect of the seed handling types (pulp consumption vs. seed regurgitation) in seed germination rates. However, pulp removal was an important factor for germination success. We conclude that T. falcklandii and X. pyrope are legitimate seed dispersers of P. lingue, since they swallow the whole fruit and regurgitate the intact seed. Abrothrix olivaceus (Waterhouse) (Cricetidae) and probably Oligoryzomys longicaudatus (Bennett) (Cricetidae) are potential secondary dispersers of P. lingue seeds, as they prefer to eat the fruit pulp rather than the seeds.
KEYWORDS: Frugivory, Austral thrush, seed germination, southern temperate rainforest.
RESUMEN
La dispersión de semillas por endozoocoría es uno de los principales mecanismos de interacción mutualista entre plantas y animales en el bosque templado de Sudamérica. En este estudio indagamos acerca de los mecanismos de dispersión primaria y estadios de depredación de semillas de Persea lingue (Ruiz et Pav.) Nees (Lauraceae), árbol endémico de estos bosques. El objetivo principal fue determinar la identidad de los vertebrados dispersores de semillas de P. lingue, distinguiendo dispersores legítimos, de consumidores de pulpa y predadores de semillas, y determinar su efecto en la germinación. Se realizaron observaciones focales de la copa de adultos de P. lingue; además, se instalaron trampas Sherman cebadas con frutos de P. lingue en fragmentos de bosque en el valle central del centro-sur de Chile. Se colectaron muestras de fecas y semillas regurgitadas, se determinó cuáles especies de roedores consumieron la pulpa, las semillas o los frutos enteros en las trampas. Se examinó si hubo remoción de frutos y semillas desde el suelo. Se realizó un ensayo de germinación para determinar si la forma de la remoción de la pulpa afecta las tasas de germinación. Registramos seis especies de aves consumiendo frutos, ninguna defecó semillas, Turdus falcklandii Quoy & Gaimard (Turdinae) y Xolmis pyrope (Kittlitz) (Tyrannidae) regurgitaron semillas intactas. Capturamos cinco especies de roedores en las trampas, cuatro dejaron semillas intactas. No encontramos efecto del tipo de manipulación de las semillas en su germinación. La remoción de la pulpa fue determinante para el éxito de germinación. Concluimos que T. falcklandii y X. pyrope son dispersores legítimos de semillas de P. lingue. Abrothrix olivaceus (Waterhouse) (Cricetidae) y probablemente Oligoryzomys longicaudatus (Bennett) (Cricetidae) son potenciales dispersores secundarios, dada su marcada preferencia por la pulpa sobre las semillas.
PALABRAS CLAVES: Frugivoría, zorzal, germinación, bosque templado lluvioso.
250 Frugivory and germination of Persea lingue (Lauraceae) sedes: GHO-ILLANEZ, D. ET AL.
INTRODUCTION al. 2005). Since has been described that P. lingue fruit pulp and seed rind exert an inhibitory effect on seed germination, Frugivory is an essential process for plant populations, probably animals seed handling may be beneficial to this especially when plant regeneration is strongly dependent plant species (Sáenz de Urtury 1986, Donoso et al. 1986, on seed dissemination by zoochory (Armesto et al. 1996, Donoso & Escobar 2006, Hernández 2007). This could Jordano 2000, Aizen et al. 2002, Cousens et al. 2008, Moran mean that a biotic vector is necessary to stimulate seed et al. 2009). Three categories of frugivory are described: germination. (1) legitimate seed dispersers: swallow the whole fruit, The specific goal was to determine which bird and defecating or regurgitating the intact seed (endozoochory); mammal species feeds on Persea lingue fruits, distinguishing (2) pulp consumers: peck fruits to obtain the pulp, dropping frugivorous quality, and to evaluate the effect of fruit and the seed; and (3) seed predators: feeds on the seeds, seed handling behaviors on seed germination rates. eliminating fruit pulp or swallowing fruits, and digesting the whole content (Jordano & Schupp 2000, Aizen et al. 2002, Bascompte & Jordano 2007). METHODS In the temperate forests of southern South America, near 70% of the tree species show a high prevalence The study site was located in an agricultural and livestock of endozoochorus dispersal syndrome, especially birds landscape composed by fragment forests and prairies. This and mammals. Along the last 25 years, there has been landscape is in a central valley of south-central Chile, in increasing interest into researching the species involved Melefquén, Panguipulli, Región de Los Ríos (72o 15’- 72o 45’ in endozoochory processes in this forest type (Armesto et W and 39o 40’- 39o 50’ S). The average annual temperature in al. 1987, Willson et al. 1996, Aizen et al. 2002, Amico & this region is about 12 °C, with a rainy season from June to Aizen 2005, Vergara et al. 2010, Mora & Soto-Gamboa September (average annual precipitation 1,800 mm, Vergara 2011, Salvande et al. 2011, Orellana et al. 2014). Many et al. 2010). The dominant tree species in the fragments of these studies described vertebrates (reptiles, birds and studied were Nothofagus obliqua, Eucryphia cordifolia, mammals) as dispersers of plant seeds. Other interactions, Aextoxicon punctatum and Laurelia sempervirens. like secondary dispersal and predation, have scarcely been During austral autumn, from April 8 to May 20, 2009, included in ecological seed studies in southern Chile, and we recorded bird feeding on 20 Persea lingue trees, few studies have included the effect of seed consumption occurring in four forest fragments that sizes approximate by vertebrates in seed germination (Bustamante et al. 1996, to 0.5 ha. The trees were located about 250 m from each Soto-Gamboa & Bozinovic 2002, Traveset et al. 2006, Reid other. We placed eight observation stations at the edge of & Armesto 2011). those four fragments, selecting visible canopies with high In the southern temperate forests, the morphology of the densities of ripe fruits. In addition, we had three observation seed dispersers matches with the size of the seeds (Armesto stations located under three P. lingue canopies, inside three et al. 1987). Few species have big fruits (more than 9.0 mm of the four forest fragments studied. One to three observers of length sensu Armesto et al. 1987); among them, there is simultaneously recorded frugivorous birds trough direct a Lauraceae, Persea lingue (Ruiz et Pav.) Nees. This tree is observations, using binoculars and a land scope. We used an endemic tree of this forests been an important component scan samplings during 60 to 120 min per session, recording of the forest association which once covered the lowlands in species identity, abundance, and residence times in the trees. the Chilean central valley. We selected this species to study In each flock, we observed one focal individual randomly its interactions with vertebrate animals due to its limited chosen, recording feeding behavior and time spent eating dispersal effectiveness as a consequence of its large sized during focal period of observation. Behavior consumption fruits that can affect its gene movement. Fruits of P. lingue categories were: a) swallowing, b) regurgitating, c) pecking ripe between March and May, after growth for 10 to 14 and d) crushing. In the 20 canopies, we did a total of 54.4 h months (Donoso & Escobar 2006, Vergara et al. 2010), and of observation during this 35-day period. represent one of the few species of the temperate rainforest We present the results as: a) visit rate: the average number with ripe fruits in autumn (Riveros & Smith-Ramírez 1996, of visits per hour for each bird species; b) accumulated Vergara et al. 2010). The fleshy fruit of this tree is a drupe, residence time: the sum of the time of permanence of each which color varies from violet to shiny black, suggesting individual on the tree, for each species, during the total bird dispersal mechanism (Armesto & Rozzi 1989, Cousens hours of observations; c) consumption rate: the sum of fruits et al. 2008). In addition, P. lingue fruits fallen by gravity can consumed by focal individuals, divided by the total hours of be eaten by understory animals. In some cases, the animals focal individuals observations for each species. move fruits away from parent plant and consume only the To record the state of seeds after bird-gut passage, we pulp leaving intact seeds. In these cases, biotic vectors, as collected bird feces under the canopy of nine Persea lingue vertebrates, become postdispersal agents (Vander Wall et trees. Under each canopy, we installed polyethylene meshes
251 Gayana Bot. 72(2), 2015 of 4 m2 of surface, 1.2 meters above the forest ground. environmental temperature, until the germination trial began Between April 12 and May 12, we collected feces, fruits and (modified from Sáenz de Urtury 1986, and Bustamante seeds for each mesh every 48 h. Collections were classified 1996). After 60 days almost without germination, 30 seeds as: nude seeds, fresh fruits and dry fruits. When possible of each treatment were exposed to solar light during eight the bird species of feces was identified. On 10 occasions hours, waiting to break the seed rind (Donoso & Escobar we collected seeds and feces immediately after direct 2006). All seeds were returned to the germination chamber observations under focal trees, from previous cleared meshes on day 61. The dependency relation among germination surfaces. This allowed us to identify the state of the seeds and treatments was evaluated with two Chi Square tests, after bird feeding. With data recorded from the meshes, we between T1 and T2, to T3 and T4. evaluated the quantity of seeds eaten by potential disperser in relation to the total seed rain, including dispersal by gravity. RESULTS To record frugivory of understory birds, such as Rhinocryptidae, and other understory animals, five trials of We recorded 138 flocks of six species visiting Persea footprint-traps of 1 x 1 meter were conducted. Traps were lingue canopies. Turdus falcklandii Quoy & Gaimard established over a flat mud surface covered with sifted flour (Turdinae) recorded higher visit frequency (65%), higher in four forest fragments. In the center of each trap we placed number of individuals (65%), and highest residence time. 20 marked seeds, 20 marked fruits, and five artificial T. falcklandii, Xolmis pyrope (Kittlitz) (Tyrannidae) and plasticine fruits. Artificial fruits had the same color and Patagioenas araucana Lesson (Columbidae) swallowed smell of P. lingue fruits. The smell was obtained by coating the whole fruit, but only T. falcklandii and X. pyrope were artificial fruits with mashed fruit pulp of P. lingue. Every observed regurgitating intact seeds. T. falcklandii ate 48 h between April 13 and May 14, we recorded the nearly double the consumption rate of X. pyrope, which is number of fruits and seeds removed and the types of marks the next important consumer of P. lingue fruits (Table I). left on the artificial fruits and sifted flour, reapplying fruit Enicognathus leptorhynchus (King) (Psittacidae) crushed pulp smell to artificial fruits. all fruits and seeds. Curaeus curaeus (Molina) (Icteridae) To record handling behavior of P. lingue fruits by and Mimus thenca (Molina) (Mimidae) pecked fruit pulp rodent, we placed 30 Sherman traps randomly distributed without removing the fruits from the branches (Table II). on the forest ground in two fragments. Inside traps we Among the 570 feces collected in the meshes, 52% were scattered 20 ripe fruits of P. lingue. Traps were installed from T. falcklandii. Its feces is approximately 3 cm long and during seven nights (May 8-15 and June 5-7), from 20:00 0.5 cm wide, commonly colored from dark blue to black, with to 08:00 (20 fruits x 30 traps x 7 nights = 420 fruit baits). white urea spots in some cases. T. falcklandii feces contained We covered the traps with litter fall to hide them, and placed countless fruit pulp remains, small whole seeds and insect wool inside to allow mammals survival during the night. fragments. 13.7% of the feces were from P. araucana, which Captured individuals were ink-marked to distinguish them is a characteristic homogeneous and round paste rounding from recaptures. In addition, in the same two fragments, in 3.0 cm of diameter, with colors that range from brown to three occasions (June 5-7) we scattered 30 ripe fruits in the violet. None of the P. araucana feces showed pulp remains, forest soil for 16 h (from nightfall to noon), distributed in seeds or another recognizable structure. The remaining 33% two samples separated by 30 meters (N=180 fruit baits). of the feces found in meshes were partially damaged or were We recorded the number of fruits with pulp removed and not recognizable. None of the feces of any species had whole missing fruits. P. lingue seeds. The effect of animal handling on germination of P. lingue From fruits and seeds collected in the meshes, 91.9% fell seeds was evaluated during a 90 day germination trial. We by gravity (N=4986), 4.7% were seeds regurgitated by birds, sowed 320 seeds in a Jacobsen chamber. Seeds locations clearly recognizable since they didn´t have pulp remains were randomly distributed on the chamber, over humid filter on their surface (N=254), and 3.3% were fruits pecked papers, and covered with transparent bells, under stable by birds, showing its pulp partially removed (N=181). In temperature of 22 oC (Figueroa 1999), and natural light ten occasions we collected seeds immediately after direct conditions. The seeds were distributed in four treatments observations under focal trees, observing high temperature with 80 seeds each one (eight replicates with 10 seeds per and drops of flesh blood upon the surface of three of the treatment): (T1) Seeds regurgitated by birds, obtained from seeds collected. the meshes; (T2) Pulp removed by rodents, obtained from On the footprint-traps we found ten rodent-type Sherman traps; (T3) Intact fruits, obtained from the meshes; footprints. Rodent teeth marks were found on six of the and (T4) Artificially removed pulp (reviewing Samuels & artificial fruits. There were no records of bird footprints. Levey 2005). All seeds were stored between 7 to 14 days in Rodents removed 52.2% of the fruits baited in those traps plastic boxes, and stratified in humid sand during 60 days at (N=94). In 79.9% of the cases rodents removed the whole
252 Frugivory and germination of Persea lingue (Lauraceae) sedes: GHO-ILLANEZ, D. ET AL. fruits (N=75), while in 13.8% the nude marked seeds were valdivianus (Philippi) (Cricetidae; N=1), Oligoryzomys removed (N=13). longicaudatus (Bennett) (Cricetidae; N=1), and Rattus We captured 29 individuals of five rodent species; three norvegicus (Berkenhout) (Muridae; N=1). In contrast, in of them were natives (Table III). We obtained an average of three nights during early June we captured 26 individuals 13.8% of capture success, with different results depending of three species: O. longicaudatus (N=18), Rattus rattus on the month of captures: in four nights during May, we (Linnaeus) (Muridae; N=5), and Abrothrix olivaceus captured three individuals of three rodent species: Geoxus (Waterhouse) (Cricetidae) (N=3).
TABLE I. Visit frequencies and residence time for bird species recorded eaten Persea lingue fruits.
TABLA I. Frecuencia de visita y tiempo de permanencia de las aves registradas consumiendo frutos de Persea lingue.
NUMBER OF BIRDS VISITS VISIT VISIT RATE ACCUMULATED RESIDENCE SPECIES FAMILY RECORDED RECORDED FREQUENCY (%) (VISITS/HOUR) TIME (HOURS)
Turdus falcklandii Turdidae 232 90 65.2 1.66 7.16 “Austral thrush” Patagioenas araucana Columbidae 42 23 16.7 0.42 2.47 “Chilean pigeon” Curaeus curaeus Icteridae 70 14 10.1 0.26 1.02 “Austral blackbird” Xolmis pyrope Tyrannidae 12 9 6.5 0.17 0.37 “Fired-eyed diucon” Enicognathus leptorhynchus Psittacidae 1 1 0.7 0.02 0.58 “Slender-billed parakeet” Mimus thenca Mimidae 1 1 0.7 0.02 0.01 “Chilean mockingbird”
TABLE II. Type of fruit handling and fruit consumption rate of bird species eating Persea lingue. Frugivory category: SD= seed disperser; PC= pulp consumer; SP= seed predator.
TABLA II. Tasa de consumo y forma de alimentación de aves consumiendo frutos de Persea lingue. Tipo de frugivoría: SD= dispersor de semillas; PC= consumidor de pulpa; SP= depredador de semillas.
AVERAGE CONSUMPTION FRUITS FRUGIVORY SPECIES FAMILY RATE PER CAPITA ECKED RUSHED CONSUMED (CATEGORY) P C WALLOWED (FRUITS/HOUR) EGURGITATED S
R (FOCAL BIRDS) Turdus falcklandii Turdidae 57 7 0 9 1.016 0.125 SD “Austral thrush” Xolmis pyrope Tyrannidae 2 2 0 1 0.800 0.098 SD “Fired-eyed diucon” Curaeus curaeus Icteridae 0 3 0 0 0.333 0.041 PC “Austral blackbird” Patagioenas araucana Columbidae 3 0 0 0 0.273 0.033 SP “Chilean pigeon” Enicognathus leptorhynchus Psittacidae 0 0 51 0 51.0 6.258 SP “Slender-billed parakeet” Mimus thenca Mimidae 0 1 0 0 1.0 0.123 PC “Chilean mockingbird”
253 Gayana Bot. 72(2), 2015
TABLE III. Fruit consumption of Persea lingue in rodents. Range: N= native; I= introduced.
TABLA III. Consumo de frutos de Persea lingue por roedores. Rango: N= nativo; I= introducido.
CONSUMPTION PREFERENCES CAPTURE CONSUMPTION SPECIES FAMILY (%) FREQUENCY (%) FREQUENCY (%) RANGE PULP WHOLE FRUIT Oligoryzomys longicaudatus Sigmodontinae N 65.5 58.2 68.9 31.1 “Long-tailed pygmy rice rat” Rattus rattus Muridae I 17.2 25.1 31.6 68.4 “Black rat” Abrothrix olivaceus Cricetidae N 10.3 13.2 80.0 20.0 “Olive-grass mouse” Geoxus valdivianus Cricetidae N 3.5 2.6 100.0 0.0 “Long-clawed mole mouse” Rattus norvegicus Muridae I 3.5 0.9 0.0 100.0 “Brown Norway rat”
A total of 227 fruits were consumed in Sherman traps seed dispersal of P. lingue seeds occurs by ingestion and by rodents. O. longicaudatus ate 132 P. lingue fruits and subsequent defecation, since in all of the bird feces sampled, seeds; picked the fruit pulp (N=91), but also the seeds intact seeds were absent. This result fits with the gape-width (N=38), and the whole fruits (N=3). Abrothrix olivaceus measurements of T. falcklandii and X. pyrope, which shows ate 30 whole fruits in the traps; picked only the fruit pulp that they are capable of feeding on large sized fleshy fruits, (N=24), but also some seeds (N=6). Rattus rattus ate but we found that they are unable to swallow them, probably 57 whole fruits; picked only the fruit pulp (N=18) and due to restrictions in their digestive tract size capability. some seeds (N=6), but mostly the whole fruits (N=33). The gape-width of T. falcklandii and X. pyrope is 10 mm Geoxus valdivianus only removed fruit pulp (N=6), and R. (Armesto et al. 1987), and the minor average width of P. norvegicus picked the whole fruits (N=2). (Table III). From lingue fruits in the study area was 10.4 mm (N=200 fruits; the 30 ripe fruits scattered in the forest soil for 16 h in June SE=0.052). Blood presence observed on the sharp top of 5 to 7 (N=180 fruit baits), we recorded 164 fruits with pulp some seeds suggest that ingestion of seeds could be harmful removed, leaving the intact seeds on forest ground (91.1%; to birds, due to the large size of the seeds in relation with SE= 0.58); and 8 whole fruits removed (4.4%; SE= 2.08). gape-width capability of frugivorous present in study During the first 59 days of germination trial, we obtained area. Curaeus curaeus was recorded as an occasional and 0.3% of germination success (N=1, from a sample of 320 selective pulp consumer. In spite of the high number of C. seeds). After solar light exposition of the seeds, we found curaeus individuals recorded per visit, they showed scarce 21.7% of the seeds germinated (N=26, from a sample of affinity for P. lingue fruits, probably due to their average 120 seeds). The germination began seven days after sun width, which is bigger than the gape-width measured for exposition. Treatment 3 (whole fruits) did not germinate this species (9.7 mm; Armesto et al. 1987) (Table II). during the 90 days of the trial, nor after sun light exposition. Turdus falcklandii showed the highest relative importance All the seeds of this treatment showed fungus proliferation. as a potential seed disperser. It was recorded feeding on We found no significant difference on germination success the fruits each day of observation, reporting more than between seeds with pulp artificially removed and seeds 50% of accumulated visits of all species, and the highest regurgitated by birds (df=1, χ2=2.16, p=0.14), nor in seeds fruit consumption rate (Tables I and II). Xolmis pyrope with pulp totally removed by rodents (df=1, χ2=1.31, p=0.14). was only occasionally recorded feeding on Persea lingue fruits, probably limited by an aggressive territorial behavior observed in T. falcklandii. P. araucana was recorded DISCUSSION swallowing fruits, but there were no records for this species regurgitating seeds, nor intact seeds in feces were collected. We found that Turdus falcklandii and X. pyrope are legitimate P. araucana probably cannot disperse any seeds, since it is dispersers of Persea lingue seeds, since they swallowed the a gizzard from the Columbidae family. Thus, we propose P. whole fruit and regurgitated intact seeds. We reject that araucana as a seed predator. The results do not agree with
254 Frugivory and germination of Persea lingue (Lauraceae) sedes: GHO-ILLANEZ, D. ET AL. the role of P. araucana as seed disperser, reported in earlier These data suggests that A. olivaceus could be an important studies (Urban 1934, Bustamante 1992, Donoso & Escobar secondary disperser of P. lingue seeds, assessing both the 2006). In those studies, the researchers reported the birds high percentage of fruits dispersed by gravity (91%), and foraging on P. lingue, but there were no reports about the the marked preference of A. olivaceus by the fruit pulp fate of seeds. Converging with our results, Amico & Aizen rather than the seeds, even at the end of fruiting season, (2005) reported P. araucana as a granivorous species. when the seeds that remained in the forest soil were scarce. Enicognathus leptorhynchus was observed in huge With this evidence, we can hypothesize that rodents that flocks of over 50 individuals predating P. lingue seeds in remove seed pulps may be a main contribute to dispersal of the same study area, but there was only one record of one P. lingue seeds. individual eating P. lingue fruits and seeds from the treetops The native frugivory species that we found feeding on sampled. We assess this species as seed predator. We did P. lingue: T. falcklandii, P. araucana, X. pyrope, M. thenca, not find evidence of understory birds as P. lingue fruit O. longicaudatus and A. olivaceus, are also described as consumers, such as the Rhinocryptidae family though they frugivorous species of Cryptocarya alba (Molina) Looser, were frequently heard in the study area. another Chilean Lauraceae tree (Bustamante et al. 1996). Rodent teeth marks on the artificial fruits show they are This result are supported by Rezende et al. (2007) study, the main fruit and seed consumer group of the forest floor which found that two phylogenetically linked species tend in the study site. We found rodents removed the pulp of the to interact with the same group of species, showing the role fruit from the ripe fruits scattered in the forest soil, during of complementarity and phylogenetic history on interaction early June. Furthermore, rodents preferred the whole fruits patterns. to naked seeds in Sherman traps. Both results suggest that Germination trial results verify the previously reported rodents are probably secondary dispersers of P. lingue seeds. inhibitory effect of the pulp (Sáenz de Urtury 1986, Donoso Data also suggest that gravity is the dominant primary & Escobar 2006), since whole fruits from Treatment 3 did dispersal mechanism for P. lingue, due to the higher not germinate and showed fungus proliferation in both pulp frequency of seeds fell collected in meshes (91.9%; and seeds. This evidence supports the positive effect of fruit N=4986), than the intact seeds regurgitated or picked by pulp removal by animal handling. birds (8.1%; N= 435). Germination trial results also indicate to being eaten Rodents showed differential affinity for P. lingue or handled by animals was not enough to stimulate throughout the season: early in P. lingue fruiting season germination of P. lingue seeds. It was necessary to expose (May) rodents showed no special affinity for P. lingue fruits, the seeds to direct solar light to begin germination. This “sun evidenced in the low capture success (N= 3), but later in shock” treatment is commonly given to P. lingue seeds in season (June) rodents showed more affinity for both fruits greenhouses to break the seed rind (Bernardo Escobar pers. and seeds, and capture success enhance (N=26). From com.). This result suggests that seed germination depends late May to June, the only species fruiting in the fragment both on biotic and abiotic factors. Previous studies (Sáenz studied was P. lingue, showing abundant fruit offer over de Urtury 1986, Donoso & Escobar 2006) found the same the forest ground in all fragments studied, due to the high results in laboratory trials with P. lingue seeds from other abundance of ripe fruits that fell under canopies by gravity. study places. With this evidence, we believe a progressive decrease of Earlier studies have found that plant species dispersed by food offer in the forest ground while winter is approaching regurgitation have a limited dispersal distance, in contrast could increase the pressure for rodents to consume seeds. with those dispersed by defecation (Cousens et al. 2008). It has been argued that secondary dispersal could be This difference is produced by the higher time of gut pass undervalued, since fruit removal is often recorded as through the animal’s digestive tract, allowing the seed to be predation, whereas seed movements could be one-step in a released by the animal farther from parent plant (Cousens et multistep seed dispersal process (Vander Wall et al. 2005). al. 2008, Vásquez 2011). In three occasions, we observed Shepherd & Ditgen (2013) found that four rodent species: and registered 7 to 10 min between ingestion of the fruit A. longipilis (Waterhouse) (Cricetidae), O. longicaudatus, and regurgitation of the seed in T. falcklandii. During this Chelemys macronyx (Thomas) (Cricetidae) and R. time, individuals were perched in nearby trees and appeared norvegicus, took Araucaria araucana (Molina) K.Koch to be waiting for the moment of regurgitation before (Araucariaceae) seeds in Argentina. The majority of these continuing foraging. This time is higher than mentioned seeds (79%) were deposited in burrows and caches in litter by Wheelwright (1991), who described the behavior soil. O. longicaudatus was seed predator and R. norvegicus of some neotropical birds that regurgitate big seeds, to deposited the seed in burrow larders with no possibilities of spent 1-2 min juggling with the seeds in their bills before germination, but A. longipilis, who removed 43% of marked dropping them. This elongated time of seed retention could seeds, left them in places where the germination was be an advantage for P. lingue seed dispersal, comparing possible. The findings are consistent with these conclusions. regurgitation mechanism versus fruit pulp handling in
255 Gayana Bot. 72(2), 2015 branches. Furthermore, in a seed dispersal study in the tree Southern South America. In: R. Lawford, P. Alaback & F. Guettarda viburnoides Cham. et Schltdl. (Rubiaceae), was Fuentes (eds.), High-Latitude Rainforests and Associated found that Cyanocorax cyanomelas Vieillot (Corvidae), a Ecosystems of the West Coast of the Americas. Climate, fruit pulp consumer, leads to positive population growing Hydrology and Conservation. pp. 248-265. Springer- Verlag New York Inc., New York, USA. of the species, whereas seed dispersal by the legitimate seed ARMESTO, J.J., R. ROZZI, P. MIRANDA & C. SABAG. 1987. Plant disperser by endozoochory, Pteroglossus castanotis Gould frugivores interactions in South American temperate (Ramphastidae), has a detrimental effects on the population forests. Revista Chilena de Historia Natural 60: 321-336. growth (Loayza & Knight 2010). BASCOMPTE, J. & P. JORDANO. 2007. Plant-Animal Mutualistic The P. lingue seed dispersal distance was documented by Networks: The Architecture of Biodiversity. Annual Vásquez (2011), in a parallel study conducted in the same Review of Ecology, Evolution, and Systematics 38: 567- study area. He found that 75-90% of P. lingue seeds are 93. dispersed no more than 10 meters away from parental trees. BUSTAMANTE, R.O. 1992. Granivoría y esparcimiento entre In spite of that, seedlings of P. lingue are commonly found plántulas y sus plantas madres: el efecto de la distancia entre plantas madres. Tesis Doctoral. Facultad de Ciencias, more than 20 m from their parent tree. In comparison, the Universidad de Chile. Santiago de Chile. 104 pp. postdispersal distance of Araucaria araucana seeds moved BUSTAMANTE, R.O. 1996. Depredación de semillas en los bosques by A. longipilis was 40 m (Shepherd & Ditgen 2013). Thus, templados de Chile. En: J.J. Armesto, C. Villagrán & we believe that probably postdispersal distance would be M.T.K. Arroyo (eds.), Ecología de los bosques nativos higher than primary dispersal distance in the case of P. de Chile (eds.), pp. 265-278. Editorial Universitaria, lingue, which seeds could be removed by A. olivaceus. Santiago, Chile. Another dispersal agent in the study site could be the wind, BUSTAMANTE, R.O., A. WALKOWIAK, C.A. HENRÍQUEZ & I. SEREY. with an average speed of 4-5 km/h from April to May, with 1996. Bird frugivory and the fate of seeds of Cryptocarya a possible maximum of 70 km/h (Webcast website 2014). alba (Lauraceae) in the Chilean matorral. Revista Chilena de Historia Natural 69: 357-363. We conclude that gravity is the main primary dispersal COUSENS, R., C. DYTHAM & R. LAW. 2008. Dispersal in Plants. A agent for Persea lingue. Turdus falcklandii is the main Population Perspective. Oxford University Press Inc., New primary disperser in bird dispersers assemblage. Abrothrix York. 232 pp. olivaceus is probably the main secondary disperser for DONOSO, C. & B. ESCOBAR. 2006. Persea lingue. En: C. Donoso P. lingue on a post dispersal stage. The effect of pulp Zegers (ed.), Las especies arbóreas de los bosques consumers in germination is positive, by pulp removing and templados de Chile y Argentina, Autoecología. Marisa consequently of seeds germination inhibition. Cuneo Ediciones, Valdivia, Chile. 678 pp. DONOSO, C., B. ESCOBAR & M. CORTÉS. 1986. Germinación de semillas y técnicas de vivero y plantaciones para las especies de los tipos forestales de la Décima Región. Informe de ACKNOWLEDGMENTS convenio Nº 102, Proyecto CONAF-UACh. 133 pp. FIGUEROA, P. 1999. Germinación de semillas de Cryptocarya Supported by ICM05-002 & PFB-23, Institute of Ecology alba (Mol.) Looser y Persea lingue Ness bajo distintas and Biodiversity (IEB). We thank two anonymous reviewers condiciones de temperatura. Tesis. Ingeniería Forestal. for valuable comments, Alejandro Miranda by his help Facultad de Ciencias Forestales, Universidad de in the field, owner of forest fragments, Denise Fliegel for Concepción. Concepción, Chile. 45 pp. english reviewing and Ngenmawida. HERNÁNDEZ, R. 2007. Análisis de un cultivo de plantas de Laurelia sempervirens, Persea lingue y Weinmannia trichosperma con dos tipos de fertilización en dos épocas de siembra. Tesis. Ingeniería Forestal. Facultad de Ciencias Forestales, REFERENCES Universidad Austral de Chile. Valdivia, Chile. 68 pp. JORDANO, P. 2000. Fruits and Frugivory. In: M. Fenner (ed.), Seeds: AIZEN, M.A., D. VÁZQUEZ & C. SMITH-RAMÍREZ. 2002. Historia The ecology of regeneration in plant communities, pp. natural y conservación de los mutualismos planta-animal 125-165. CABI Publications, Wallingford, U.K. del bosque templado de Sudamérica austral. Revista JORDANO, P. & E. SCHUPP. 2000. Seed Disperser Effectiveness: the Chilena de Historia Natural 75: 79-97. quantity component and patterns of seed rain for Prunus AMICO, G.G. & M.A. AIZEN. 2005. Dispersión de semillas por aves mahaleb. Ecological Monographs 70(4): 591-615. en un bosque templado de Sudamérica austral: ¿quién LOAYZA, A.P. & T. KNIGHT. 2010. Seed dispersal by pulp consumers, dispersa a quién? Ecología Austral 15: 89-100. not ‘‘legitimate’’ seed dispersers, increases Guettarda ARMESTO, J.J. & R. ROZZI. 1989. Seed dispersal syndromes in viburnoides population growth. Ecology 91(9): 2684- the rain forests of Chiloé: evidence for the importance 2695. of biotic dispersal in a temperate rain forest. Journal of MORA, J.P. & M. SOTO-GAMBOA. 2011. Legítima dispersión de Biogeography 16: 219-226. semillas Ugni molinae Turcz (Myrtaceae) por monito del ARMESTO, J.J., C. SMITH-RAMÍREZ & C. SABAG. 1996. The Importance monte, Dromiciops gliroides. Gayana Botánica 68(2): 309- of Plant-Bird Mutualisms in the Temperate Rainforest of 312.
256 Frugivory and germination of Persea lingue (Lauraceae) sedes: GHO-ILLANEZ, D. ET AL.
MORAN, C., C. CATTERALL & J. KANOWSKI. 2009. Reduced dispersal two mechanisms of fruits processing on seed germination. of native plant species as a consequence of the reduced Plant Ecology 159: 171-174. abundance of frugivore species in fragmented rainforest. TRAVESET, A., A.W. ROBERTSON & J. RODRÍGUEZ-PÉREZ. 2006. A Biological Conservation 142: 541-552. review on the role of endozoochory in seed germination. MURÚA, R. 1996. Comunidades de mamíferos del bosque templado In: A.J. Dennis, E.W. Schupp, R.J. Green & D.W. de Chile. En: J.J. Armesto, C. Villagrán & M.T.K. Arroyo Westcott (eds.), Seed dispersal: theory and its application (eds.), Ecología de los bosques nativos de Chile, pp. 113- in a changing world, pp. 78-101. CAB International, 133. Editorial Universitaria, Santiago, Chile. Wallingford, UK. ORELLANA, J.I., C. SMITH-RAMÍREZ, J.R. RAU, S. SADE, A. GANTZ URBAN, O. 1934. Botánica de las plantas endémicas de Chile. & C.E. VALDIVIA. 2014. Sincronía fenológica entre el Sociedad de Impresores Literarios de Concepción. zorzal austral Turdus falcklandii (Turdidae) y sus recursos Concepción, Chile. 289 pp. alimenticios en bosques y praderas del sur de Chile. VANDER WALL, S.B., K.M. KHUN & M. BECK. 2005. Seed removal, Revista Chilena de Historia Natural 87: 11. seed predation and secondary dispersal. Ecology 86(3): REID, S. & J.J. ARMESTO. 2011. Avian gut-passage effects on seed 801-806. germination of shrubland species in Mediterranean central VÁSQUEZ, I. 2011. Patrones espaciales de reclutamiento y dispersión Chile. Plant Ecology 212: 1-10. de semillas del árbol Persea lingue (Miers ex Bertero) REZENDE, E., P. JORDANO & J. BASCOMPTE. 2007. Effects of Ness (Lauraceae) en el bosque del valle del sur de Chile, phenotypic complementarity and phylogeny on the nested efectos del hábitat y un corredor. Tesis. Ingeniería en structure of mutualistic networks. Oikos 116: 1919-1929. Recursos Naturales. Facultad de Ciencias Agronómicas, RIVEROS, M. & C. SMITH-RAMÍREZ. 1996. Patrones de floración Universidad de Chile. Santiago, Chile. 47 pp. y fructificación en bosques del sur de Chile. En: J.J. VERGARA, P., C. SMITH, C. DEL PIANO, I. ORELLANA, D. GHO & I. Armesto, C. Villagrán & M.T.K. Arroyo (eds.), Ecología VÁSQUEZ. 2010. Frugivory on Persea lingue in temperate de los bosques nativos de Chile, pp. 235-249. Editorial Chilean forest: interactions between fruit availability Universitaria, Santiago, Chile. and habitat fragmentation across multiple spatial scales. SÁENZ DE URTURY, M.X. 1986. Estudio anatómico y de germinación Oecologia 164: 981-991. en lingue (Persea lingue Nees). Tesis. Ingeniería Forestal. WEBCAST WEBSITE 2014. Panguipulli Metereological Station Facultad de Ciencias Agrarias y Forestales, Universidad de downloaded on May 2014
Recibido: 09.12.14 Aceptado: 24.08.15
257